Effect of UV-A Light on the Chaperone-like Properties of Young and Old Lens α-Crystallin

Abstract

purpose. To study the damaging effect of UV-A irradiation on the chaperone-like properties of α-crystallin and the subsequent recovery process of young and old bovine lenses.

methods. Young and old bovine lenses were kept in organ culture. After 24 hours of incubation they were irradiated with UV-A at 365 nm, and optical quality measurements were performed during the experiments (192 hours).α -Crystallin and αA1-, αA2-, αB1-, and αB2-crystallin subunits were analyzed, separated by gel filtration and cation exchange chromatography, respectively, after different culture times. Protein patterns were obtained after two-dimensional (2-D) gel electrophoresis. Chaperone-like activity was determined on the basis of insulin B-chain and βL-crystallin aggregation assays. Aggregation of α-crystallin was analyzed, tryptophan fluorescence measurements were performed, andα -crystallin mRNA levels were determined.

results. The water-soluble α-crystallin obtained from old lenses compared with young lenses after UV irradiation had decreased chaperone activity, a higher molecular weight, and increased loss of tryptophan fluorescence. Moreover, α-crystallin mRNA virtually disappeared, whereas extra spots on the 2-D protein pattern appeared, possibly because of deamidation.

conclusions. α-Crystallin obtained from old lenses is more affected by irradiation than α-crystallin derived from young lenses. Moreover, it appeared that αB-crystallin from UV-treated old lenses compared with control lenses was less susceptible to UV-A than αA-crystallin. It may well be that αB-crystallin protects αA-crystallin in vivo.

The two subunits of the 800-kDa α-crystallin aggregate, a major structural eye lens component, are αA-crystallin andα B-crystallin. ¹ Not long ago, it became apparent that these subunits are not lens specific, because they also exist in a variety of other tissues. ^{2 3} Both αA- andα B-crystallin belong to the family of small heat shock proteins. ⁴ Similar to other heat shock proteins αA- andα B-crystallin also have chaperone-like activity, which means that they can assist in refolding of denatured proteins and function in the prevention of undesired protein association induced by stress conditions. ⁵  

Solar radiation is believed to be one of the major environmental stress factors involved in cataract formation and may be involved in senile cataractogenesis too. ^{6 7 8 9} Recent studies ¹⁰ have shown that during aging and cataract formation, there are some major posttranslational modifications of αA- and αB-crystallin that alter their chaperone properties. Conditions that are assumed to play a role in these modifications are oxidation, truncation at the C-terminal region, racemization, phosphorylation of various serines, and deamidation of Gln and Asn. Studies performed previously show that exposure of isolated α-crystallin to UV-B light at different wavelengths between 280 and 308 nm is associated with gradual loss of chaperone–protein efficacy. ^{11 12 13 14} Furthermore, it has been shown that exposure of bovine lenses to UV-A radiation in long-term organ culture has a damaging effect on lens enzymes and proteins. ^{15 16 17}  

In the present article, we describe the damaging effect of UV-A irradiation at 365 nm on the individual subunits of α-crystallin and the subsequent recovery process of young and old bovine lenses in long-term organ culture as monitored by laser scanning of the optical quality. To extend our knowledge of UV-A–related modifications ofα -crystallin acquired from previous studies, ¹⁶ we analyzed α-crystallin and αA- and αB-crystallin subunits after irradiation of the cultured lenses. 

The modifications of αA- and αB-crystallin, paralleled by the changes of the chaperone-like activity of the subunits after irradiation were examined and compared with the change of chaperone-like activity of native α-crystallin. Lens mRNA was studied also to verify whether the observed changes occur at the level of translation or transcription. 

Methods

Irradiation of Lens in Organ Culture

Lenses were carefully excised from 1-year-old and 2- to 4-year-old bovine eyes. For present purposes, lenses a maximum of 1-year-old will be defined as “young” and those 2 to 4 years old as “old.” After inoculation (2–4 hours) each lens was placed in a specially designed culture system and irradiated as described previously. ^{15 16} The lenses received 33 J/cm² of 365 nm energy when exposed for 75 minutes. The 400-W UV lamp contained a filter that provided radiation of 8.5 mW/cm² at 365 nm. ¹⁶ The radiation was 7.465 mW/cm², measured by an IL 1700 Radiometer (International Light, Newburyport, MA). The temperature of the culture dish did not exceed 37°C. After irradiation, the optical quality of the lens was monitored throughout the culture period. Optical measurements were performed as described earlier. ^{15 18} For each time interval of control and irradiated lenses we used four young and four old lenses, respectively. 

Preparation of Lens Extract and Purification of α-Crystallin andα A1-, αA2-, αB1-, and αB2-Crystallin Subunits

Lenses were dissected under a binocular stereomicroscope. A cut along the equator was made, the epithelium was removed, and the lens was cut into three parts: cortex, equator, and nucleus. The equator and the nucleus were immediately stored at −20°C for other assays. Lens cortex was homogenized in 100 mM Tris buffer at pH 7.5 and spun at 4°C in an Eppendorf (Freemont, CA) tube at 13,000 rpm for 30 minutes. The supernatant is the water-soluble fraction. The pellet was stored in 5 M urea solution to be examined in forthcoming studies. Separation of the water-soluble fraction into α-, βH-, βL-, and γ-crystallin was performed by gel filtration on Sephacryl S-300 HR (Pharmacia-LKB, Uppsala, Sweden). ¹⁹ The column was loaded with 100 mg/ml water-soluble lens protein, and the separated fractions were measured automatically at 280 nm. Determination of the aggregation size ofα -crystallin fractions from control and UV-irradiated lenses was performed by comparing elution times on a Superose 6 HR column (Pharmacia-LKB) in 20 mM sodium phosphate and 100 mM Na₂SO₄ (pH 6.9) at a flow rate of 0.5 ml/min. The relative scattering was monitored at 280 nm. αA1-, αA2-,α B1-, and αB2-crystallin subunits were obtained from α-crystallin by cation exchange chromatography on a CM-52 carboxymethyl cellulose (Whatman, Clifton, NJ) column (1.5 × 20 cm) at 4°C, using a gradient buffer ranging from 0.04 M to 0.2 M NaAC in 8 M urea (pH 5.0) at a flow rate of 0.5 ml/min. ²⁰ Protein concentration was determined using BCA protein assay reagents (Pierce, Rockford, IL). 

Mini One- and Two-Dimensional Gel Electrophoresis

Mini one-dimensional gel electrophoresis (13% sodium dodecyl sulfate) was performed under conditions described by Laemmli. ²¹ Mini two-dimensional (2-D) polyacrylamide gel electrophoresis was performed essentially according to the method of O’Farrell ²² for large gels. Minor modifications were made as described previously. ²³ Analysis of the resultant Coomassie blue–stained 2-D gels was performed using a densitometer (Master-scan; Scanalytics, Billerica, MA) . 

Insulin B-Chain and βL-Crystallin Aggregation Assays

The chaperone-like activity of α-crystallin and αA1-,α A2-, αB1-, and αB2-crystallin subunits from control and UV-irradiated lenses, was determined by two different assays. Heat-induced aggregation assay of βL-crystallin was performed at 60°C as described by Horwitz. ²⁴ The proteins were dissolved in 20 mM sodium phosphate, 100 mM Na₂SO₄, and 10 mM EDTA (pH 6.9). The scattering was recorded in a spectrophotometer (Lambda 2UV/VIS; Perkin–Elmer, Norwalk, CT) at 360 nm for 30 minutes. Heat-induced aggregation of the insulin B-chain was performed at 360 nm, 37°C for 20 minutes, as described by Horwitz et al. ⁵ The proteins were dissolved in 20 mM sodium phosphate, 100 mM Na₂SO₄, and 10 mM EDTA (pH 6.9). At time 0 of the measurements 0.2 mM dithiothreitol was added. Both assays were performed in duplicate with a concentration of 250μ g/ml substrate proteins and 100 and 200 μg/ml α-crystallin andα A1-, αA2-, αB1-, and αB2-crystallin subunits. 

Tryptophan Fluorescence

Tryptophan fluorescence measurements were preformed with 100μ g/ml protein in a 1-ml solution of 20 mM sodium phosphate and 100 mM Na₂SO₄ (pH 6.9), with a fluorescence spectrophotometer (model 650-40; Perkin–Elmer). The excitation wavelength was set to 295 nm, and the fluorescence emission was detected at 330 nm. 

RNA Isolation and Northern Blotting

Isolation of total RNA from control and irradiated bovine lenses was performed by reagent assay (Trisol; Gibco, Grand Island, NY). Twenty micrograms total RNA was denatured with 6 M glyoxal and 50% dimethyl sulfoxide in 0.1M sodium phosphate buffer (pH 7.0) for 60 minutes at 50°C. The glyoxalated RNA was transferred immediately after electrophoresis from a 1% agarose gel to pure membrane (Hybond N+; Amersham, Amersham, UK) by capillary elution using 20× SSC buffer. Northern blot analyses were hybridized according to Church and Gilbert, ²⁵ by using αA- and αB-crystallin cDNA hybridization mixtures. To confirm that equal amounts of RNA were loaded in each lane, the blots were stripped and hybridized afterward to an 18S ribosomal probe. 

Results

Young and old bovine lenses were kept in organ culture. After 24 hours of incubation they were irradiated with UV-A at 365 nm, and optical quality measurements were performed during the 8 days (192 hours) of the experiments. The focal length of the position of the beams passed through young and old lenses during the scans is depicted in Figure 1 . The focal length represents the focus of the lens. There is no change of this parameter in control lenses during the culture time. However, it was shown that 24 hours after irradiation there were some changes of the focal length of both young and old lenses that returned to control levels 24 hours later. Ninety-six hours after irradiation, the focal length changed again, but apparently more in old than in young lenses. These results led us to restrict the measurements to control and UV-A–treated young and old lenses that were in the culture 48, 96, and 120 hours, (24, 72, and 96 hours after irradiation), respectively. 

Separation of water-soluble α-crystallin from control and UV-irradiated young and old lenses was performed on a Sephacryl S-300 gel filtration column. The fractions of α-crystallin were collected and analyzed by mini 1- and 2-D polyacrylamide gel electrophoresis (2-D PAGE). UV-A–induced modifications were observed only in the 2-D PAGE pattern, as shown in Figures 2 and 3 . The damaging effect of UV-A irradiation on water-solubleα A-crystallin is reflected by extra spots (arrows in Fig. 2b ), which were seen and scanned 24 hours after irradiation. These spots had the same molecular weight as αA-crystallin in young lenses. In addition, in old lenses there were spots with the same molecular weight as αA- and αB-crystallin (arrows in Figs. 3b and 3c ). In view of the localization and migration distance of the extra spots, they may be due to deamidation of αA- and αB-crystallin components, as is the case in chicken lenses. ²⁶ Although these extra spots still existed in old lenses 72 hours after irradiation (Fig. 3c) , in young lenses they were not observed at this time point (compare Fig. 2c ). 

The chaperone properties of the purified water-soluble α-crystallin fractions were determined by insulin B-chain and βL-crystallin aggregation assays with different concentrations of α-crystallin. Our results show that α-crystallin from old bovine lenses possesses only a decreased chaperone-like activity, in agreement with other studies described previously. ²⁷ There is no significant difference between α-crystallin activity detectable from control and UV-irradiated young lenses (not shown). Figure 4 illustrates the chaperone-like activity of different concentrations of the water-soluble α-crystallin from old control and UV-irradiated lenses, which was performed with the insulin assay (Figs. 4a and 4c) and the heating assay with βL-crystallin (Figs. 4b and 4d) . The results are summarized in Figure 4e . A decrease of 20% of chaperone activity occurred (with 100 and 200 μg α-crystallin) 72 hours after irradiation. Twenty-four hours later, the chaperone ability recovered, measured by insulin assay. In the heating assay, the chaperone activity of α-crystallin decreased by 30% with 200 μg α-crystallin and by 55% with 100 μg protein at 72 hours after irradiation and slightly recovered 24 hours later by 10% and 20%, respectively. 

Tryptophan fluorescence measurements were performed with α-crystallin fractions from control and irradiated young and old lenses to verify whether conformational changes might have occurred (Fig. 5) . α-Crystallin from young lenses showed decreased values of tryptophan fluorescence 24 hours after irradiation and returned to the control level 96 hours after irradiation. α-Crystallin from old lenses showed up to a 2.5-fold loss of tryptophan fluorescence as a function of time after irradiation and only a slight recovery 96 hours after irradiation. In view of the these results, further experiments were restricted to old lenses. 

Determination of the aggregation size of α-crystallin from control and UV-irradiated old lenses was performed by comparing elution times on a Superose 6 HR column (Fig. 6) . α-Crystallin aggregates from UV-treated lenses (72 and 96 hours after irradiation) seemed to be larger than in the control samples. 

αA1-, αA2-, αB1-, and αB2-crystallin subunits were obtained from the water-soluble α-crystallin of old 96-hour cultured lenses (72 hours after irradiation) after cation exchange chromatography on a CM-52 carboxymethyl cellulose column (data not shown). The purity of the yielded fractions was approximately 95% as shown previously. ²⁰  

The chaperone properties of the purified subunits of α-crystallin were determined by insulin B-chain and βL-crystallin aggregation assays, as shown in Figure 7 . Both phosphorylated αA- and αB-crystallin subunits (αA1, αB1) seem to be better chaperones than nonphosphorylated subunits (αA2,α B2) derived from control and irradiated lenses. Measurements with the heating assay (Figs. 7b and 7d) indicate that αA-crystallin from control lenses provided better protection than αB-crystallin, in agreement with a previous study. ²⁸ With the insulin assay, however, both αA- and αB-crystallin from control lenses provided protection (Figs. 7a and 7c) . Comparison of αB-crystallin toα A-crystallin chaperone activity in control lenses with that of irradiated lenses, as measured by the insulin assay and the heating assay, shows that αB-crystallin from UV-treated lenses was a better chaperone than αA-crystallin. αA-crystallin from irradiated lenses lost almost all its chaperone activity, measured with the heating assay. 

To check whether UV damage occurs at the level of transcription or mRNA degradation, total RNA was isolated from control and irradiated old bovine lenses, followed by hybridization with αA- and αB-crystallin cDNA probes. Using an rRNA probe as a control (Fig. 8 a), it appeared that αA- and αB-crystallin mRNA degraded at 72 hours after irradiation, whereas 24 hours later synthesis started again (Figs. 8b and 8c) . 

Discussion

We investigated in situ the effect of UV-A irradiation on the chaperone-like ability of α-crystallin and its subunits from young and old cultured bovine lenses to create similar conditions and effects in the lens as those that take place during solar radiation. 

In a comparison of untreated with UV-A–treated lenses, it appeared that the water-soluble fractions of α-crystallin obtained from old lenses were more affected than α-crystallin derived from young lenses. For instance, such effects were: decreased ability ofα -crystallin to inhibit protein denaturation in vitro, higher molecular weight of the α-crystallin fractions, loss of tryptophan fluorescence, degradation of α-crystallin mRNA, and the appearance of extra spots on the 2-D pattern, possibly because of deamidation. These phenomena were paralleled by a higher damage of the optical quality of old lenses compared with young lenses. But at the time that lenses started to recover, measured by the focal length repair, the chaperone-like activity recovered, tryptophan fluorescence increased, and the extra spots with the same subunit molecular weight of αA- andα B-crystallin disappeared. Previous work ¹⁵ has shown that some metabolic enzymes also can recover from UV-A damage. The increase in αA- and αB-crystallin mRNA levels may play a role in the recovery, because de novo synthesis of α-crystallin can occur in the outer cortex of the lenses. 

Studies of the effect of UV-B on αA-crystallin ²⁹ have shown specific racemization and isomerization after irradiation. Our results show that αA-crystallin from old lenses seemed to be more susceptible to UV-A than αB-crystallin, which provided a better protection to denatured proteins. The differences in results obtained with the insulin and the heating assay may be due to αA-crystallin’s inability to function properly as a chaperone. It appeared from this study and from other work ²⁸ that αB-crystallin was temperature sensitive and therefore had less chaperone capacity at elevated temperatures than αA-crystallin. In contrast,α B-crystallin from UV-treated lenses compared with control lenses was a better chaperone than αA-crystallin, which lost almost all its chaperone-like activity after irradiation. It may well be thatα B-crystallin and αA-crystallin protect each other as the subunits interact in vivo. Moreover, our results are consistent with previous work, ²⁸ because phosphorylated αA- and αB-crystallin subunits (αA1, αB1) are better chaperones than the nonphosphorylated subunits (αA2, αB2) from control and UV-A irradiated lenses. 

Identification of the photo-oxidation sites (UV-B) in bovineα -crystallin ³⁰ indicates that the N-terminal regions ofα A- and αB-crystallin are exposed to an aqueous environment and are in the vicinity of tryptophan residues from neighboring subunits. Our study shows a decrease of tryptophan fluorescence of isolated bovineα -crystallin from lenses exposed to UV-A light. Posttranslation modifications and formation of high molecular aggregates of αA-andα B-crystallin after UV-A irradiation can lead to changes of the chaperone-like properties that may affect the integrity of other key proteins involved in the structure of the cytoskeleton. In turn, this may result in opacification of the lens during aging. 

 Supported by a Marie Curie Research Training Grant in Biotechnology from the European Commission for Science, Research, and Development (Bio. 4-CT96-5121) and in part by the Matrix Biology Institute, Ridgefield, New Jersey.

 Submitted for publication January 7, 1999; revised April 23 and June 28, 1999; accepted July 28, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Composite of average focal length over time in organ culture for young and old bovine lenses after UV-A irradiation of 33 J/cm² compared with control lenses.

Figure 2.

 

Mini 2-D PAGE of water-soluble α-crystallin fraction obtained from the cortex of cultured young bovine lenses. (a) Protein profile of control lens after 120 hours’ culturing, (b) after 48 hours’ culturing (24 hours after irradiation), (c) after 96 hours’ culturing (72 hours after irradiation), and (d) after 120 hours’ culturing (96 hours after irradiation). Arrows: extra spots. IEF, isoelectric focusing.

Figure 3.

 

Mini 2-D PAGE of water-soluble α-crystallin fraction obtained from the cortex of cultured old bovine lenses. (a) Protein profile of control lens after 120 hours’ culturing, (b) after 48 hours’ culturing (24 hours after irradiation), (c) after 96 hours’ culturing (72 hours after irradiation), and (d) after 120 hours’ culturing (96 hours after irradiation). Arrows: extra spots. IEF, isoelectric focusing.

Figure 4.

 

Chaperone-like activity of water-soluble α-crystallin fraction obtained from the cortex of control and UV-A–irridiated old bovine lenses determined by two different assays and recorded in a spectrophotometer at 360 nm. (a) Insulin assay at 37°C with 250 μg insulin and 100 μg α-crystallin. (b) Heating assay at 60°C with 250 μg βL-crystallin and 100 μgα -crystallin. (c) Insulin assay with 250 μg insulin and 200 μg of α-crystallin. (d) Heating assay with 250 μgβ L-crystallin and 200 μg α-crystallin. Curve A: substrate protein alone; curve B: plus α-crystallin obtained from UV-treated lenses 72 hours after irradiation; curve C: plus α-crystallin obtained from UV-treated lenses 96 hours after irradiation; curve D: plusα -crystallin obtained from UV-treated lenses 24 hours after irradiation; curve E: plus α-crystallin obtained from control lenses. The results are summarized in (e). Error bars, SD of 4 different measurements each.

Figure 5.

 

The relative tryptophan fluorescence (excitation at 295 nm/emission at 330 nm) of water-soluble α-crystallin fraction obtained from control and UV-A–irridiated young and old bovine lenses. Error bars, SD of four different measurements each.

Figure 6.

 

Elution profile of comparative gel permeation chromatography on a Superose 6 HR column to determine the aggregation size of water-soluble α-crystallin fractions from control and UV-A–irridiated old bovine lenses. The relative scattering was monitored at 280 nm.

Figure 7.

 

Chaperone-like activity of αA1-, αA2-, αB1-, and αB2-crystallin subunits of control and UV-A–irridiated old bovine lenses determined by two different assays and recorded by spectrophotometer at 360 nm. (a) Insulin assay at 37°C with 250 μg insulin and 100μ g αA-crystallin. (b) Heating assay at 60°C with 250μ g βL-crystallin and 100 μg αA-crystallin. Curve A: substrate protein alone; curve B: plusα A2-crystallin obtained from UV-treated lenses 72 hours after irradiation; curve C: plus αA1-crystallin obtained from UV-treated lenses 72 hours after irradiation; curve D: plus αA2-crystallin obtained from control lenses; curve E: plus αA1-crystallin obtained from control lenses. (c) Insulin assay at 37°C with 250 μg insulin and 100μ g αB-crystallin. (d) Heating assay at 60°C with 250μ g βL-crystallin and 100 μg αB-crystallin. Curve A: substrate protein alone; curve B: plusα B2-crystallin obtained from UV-treated lenses 72 hours after irradiation; curve C: plus αB1-crystallin obtained from UV-treated lenses 72 hours after irradiation; curve D: plus αB2-crystallin obtained from control lenses; curve E: plus αB1-crystallin obtained from control lenses. The results are summarized in (e). Error bars, SD of four different measurements each.

Figure 8.

 

Northern blot analysis of total RNA from control and UV-A–irridiated old bovine lenses probed for (a) 18S rRNA to equalize density measurements and then stripped and rehybridized with (b) αA-crystallin and (c) αB-crystallin cDNA probes.

The authors thank Wilfried de Jong for valuable advice and Anke van Rijk, Perry Overkamp, and Neil Azzam for technical assistance.